of F430 in Active Methyl-Coenzyme M Reductase - American

Coenzyme B Induced Coordination of Coenzyme M via Its Thiol Group to Ni(I) of F430 in Active Methyl-Coenzyme M Reductase. Cinzia Finazzo,† Jeffrey ...
6 downloads 0 Views 67KB Size
Published on Web 04/05/2003

Coenzyme B Induced Coordination of Coenzyme M via Its Thiol Group to Ni(I) of F430 in Active Methyl-Coenzyme M Reductase Cinzia Finazzo,† Jeffrey Harmer,† Carsten Bauer,‡ Bernhard Jaun,‡ Evert C. Duin,§,| Felix Mahlert,§ Meike Goenrich,§ Rudolf K. Thauer,§ Sabine Van Doorslaer,†,⊥ and Arthur Schweiger*,† Physical Chemistry and Organic Chemistry, ETH Zurich, CH-8093 Zurich, Switzerland, and Max-Planck-Institut fu¨ r terrestrische Mikrobiologie, D-35043 Marburg, Germany Received January 31, 2003; E-mail: [email protected]

Methane is formed in methanogenic archaea by the reduction of methyl-coenzyme M (CH3-S-CoM) with coenzyme B (HS-CoB) to CH4 and the heterodisulfide CoM-S-S-CoB.1 This reaction is catalyzed by methyl-coenzyme M reductase (MCR), which is composed of three different subunits in an R2β2γ2 arrangement and which contains tightly bound 2 mol of the nickel porphinoid F430. Crystal structures of the 300 kDa enzyme with and without coenzymes or product bound have been resolved to 1.16 Å.2 They were, however, only obtained for the enzyme in the inactive Ni(II) state. For the enzyme to be active, the prosthetic group has to be in the Ni(I) oxidation state,3-5 which is rapidly lost by autoxidation of the Ni(I).6 Therefore, it has not been known until now how within the active enzyme the active site Ni(I) interacts with the substrates. Active MCR exhibits an axial EPR signal MCRred1 derived from Ni(I) which does not change significantly when CH3-S-CoM alone or together with HS-CoB are added to the active enzyme, showing that although an enzyme-substrate-complex is formed, there is no detectable direct interaction of the substrates with the Ni(I) of the prosthetic group.7 The XAS data of the active enzyme are also the same in the absence and presence of the substrates.8 The reduction of CH3-S-CoM catalyzed by MCR is inhibited by coenzyme M (HS-CoM), inhibition being reversible and competitive to CH3-S-CoM. In the presence of only HS-CoM, the enzyme shows the axial MCRred1 EPR signal. In the presence of both HS-CoM and HS-CoB, however, the axial signal is partially converted into the highly rhombic EPR signal MCRred2.7 On the basis of 1H and 14N data obtained from electron nuclear double resonance (ENDOR) and hyperfine sublevel correlation spectroscopy (HYSCORE) measurements, it was proposed that in the MCRred2 state HS-CoM is axially coordinated to Ni(I).9 Here we report experiments with 33S-labeled HS-CoM, proving that the thiol group of HS-CoM coordinates to the Ni(I) ion of F430. [2-33S]-coenzyme M (4) (Scheme 1) was synthesized in a onepot procedure starting from elemental sulfur ([33S8]-S8)10 and potassium cyanide11 according to Scheme 1.12 Figure 1 shows the X-band EPR spectra of MCRred2-HS-CoM (32S (99.25%) with nuclear spin I ) 0, 33S (0.75%) with I ) 3/2) and MCRred2-H33S-CoM. For a better comparison, the signals of MCRred1 were subtracted from the red1/red2 mixture normally shown by these preparations. The EPR spectrum of 33S-labeled MCRred2 shows a pronounced line broadening at the high-field feature corresponding to the g3 principal value. This is a strong indication for the presence of a large 33S hyperfine coupling along this principal axis direction. No significant broadenings are observed † Physical Chemistry, ETH Zurich. ‡ Organic Chemistry, ETH Zurich. § Max-Planck-Institut. | Current address: Department of Chemistry,

Auburn University, AL 368495312. Current address: Spectroscopy in Biophysics and Catalysis (SIBAC Laboratory), University of Antwerp, 2610 Wilrijk, Belgium. ⊥

4988

9

J. AM. CHEM. SOC. 2003, 125, 4988-4989

Scheme 1. Synthesis of [2-33S]-Coenzyme Ma

a Conditions: (a) N , KCN, EtOH, 4 h, reflux. (b) N , BrCH CH SO Na, 2 2 2 2 3 DMF, 4 h, 120 °C. (c) (1) N2, K2CO3, H2O, 1 h, 60 °C; (2) H+, NH3. (d) 33 N2, DTT, H2O, 30 min. Overall yield from [ S8]-S8, 87% (NMR); after final purification for MCR-assay, 44%.

Figure 1. X-band EPR spectra of MCR in the MCRred2 state (coenzyme M inhibited enzyme in the presence in coenzyme B). (a) With HS-CoM. (b) With H33S-CoM. Solid lines: experimental spectra after subtraction of MCRred1 signals. Dashed lines: simulations, spectrum b used the same spin Hamiltonian parameters as in (a) but with the addition of a 33S hyperfine interaction, |A1,2| ) 20 MHz, |A3| ) 35 MHz. Experimental conditions: 77 K, modulation amplitude 0.6 mT, microwave frequency 9.45 GHz (Supporting Information contains further details).

at g1 and g2. From spectral simulations, the 33S hyperfine coupling along g3 is estimated to be roughly |A3| ) 35 MHz, with upper limits along g1 and g2 of |A1,2| ) 25 MHz. In contrast to MCRred2, when either the ox1 or the red1 form of MCR is incubated in the presence of H33S-CoM, no significant line broadenings are observed in the EPR spectra. This shows that any interaction with 33S in these two forms is small (as compared to the EPR spectral resolution). A clear-cut proof of the coordination of HS-CoM to Ni(I) is obtained from HYSCORE spectra measured at Q-band13 at the lowfield end (g1 value) of the EPR spectrum. Figure 2a and 2b shows the single-crystal-like HYSCORE spectra of MCRred2-HS-CoM and MCRred2-H33S-CoM at g1 which are free of contributions from other paramagnetic species of MCR. The additional peaks observed in Figure 2b originate from 33S interactions (labeled in Figure 2b). The two cross-peaks in the (- +)-quadrant at (-10.8, 31.8) MHz and (-31.8, 10.8) MHz are assigned to triple-quantum transitions 10.1021/ja0344314 CCC: $25.00 © 2003 American Chemical Society

COMMUNICATIONS

Figure 2. Q-band HYSCORE spectra recorded at 25 K of MCR in the MCRred2 state; observer position at g1. (a) With HS-CoM. (b) With H33SCoM. The arrows identify peaks that originate from 33S interactions. The Supporting Information contains further details.

with ∆mI ) 3. For a simplified system with an isotropic g tensor and an axial hyperfine tensor, the two triple-quantum frequencies can be written to first order as

[(

)

(

)

2 2 A⊥ A| + νI sin2 β + ( + νI cos2 β 2 2

ν(+/-) )3 ( TQ

]

1/2

with the hyperfine principal values A⊥ and A|, the nuclear Zeeman frequency νI, and the angle β between the A| principal axis and the static magnetic field vector B0. For a nuclear quadrupole interaction that is small as compared to the hyperfine interaction, these frequencies are to first-order independent of the nuclear quadrupole interaction and differ by 6νI for β ) 0, 90°. In Figure 2b, the observed splitting of 20.9 MHz is slightly smaller than 6νI ) 21.7 MHz, indicating that the orientations selected in this experiment are close to the principal axis of A⊥. The 33S hyperfine coupling A along g1 can easily be estimated from the equation 2

2

(-) 2 ν(+) TQ - νTQ ) 18νI(aiso + T(3 cos β - 1)) ) 18νIA

where aiso is the isotropic hyperfine coupling, and T is the dipolar coupling.14 Along g1, we then find |A| ) 13.8 MHz, and for the principal value |A⊥| we estimate a coupling of about 15 MHz. Several additional peaks are observed in the (+ +)-quadrant of Figure 2b. The strong diagonal peak at 23 MHz is most probably a sulfur double-quantum transition (∆mI ) 2), and the cross-peaks represent correlations between sulfur transitions and/or nitrogensulfur combination transitions. An unequivocal assignment of all of the new peaks observed in the 33S sample is difficult because the HYSCORE spectrum could only be observed along g1. This is because the large anisotropy of the 33S hyperfine coupling broadens the peaks beyond detection as soon as the B0 observer field in the HYSCORE experiments is shifted to higher values.

The combination of EPR and HYSCORE data proves that, in the MCRred2 state, HS-CoM is directly coordinated to the Ni(I) ion. An estimate of the spin density on 33S can be obtained from the hyperfine tensor and by considering the relative signs of the principal values. Assuming three positive principal values [(A1, A2, A3) ) (15, 15, 35) MHz] yields an isotropic part of aiso ) 21.7 MHz and a dipole part of (-6.7, -6.7, 13.4) MHz. For a hyperfine tensor [(A1, A2, A3) ) (-15, -15, 35) MHz], the isotropic part is aiso ) 1.7 MHz, and the dipole part is (-16.7, -16.7, 33.4) MHz. In the first case, a spin density of 0.6% in the s-orbital is estimated from the isotropic part, and a spin density of 7% in a 3p-orbital is estimated from the dipolar part.15 For the second case, the corresponding values are 0.05% (s-orbital) and 17% (3p-orbital). In either case, the large spin density on the sulfur ligand is further proof that the ground state of MCRred2 has a relatively high percentage of dz2 character, and the large hyperfine coupling A3 corroborates the proposal that the g3 principal axis is perpendicular to the macrocycle.9 On the basis of the finding that the HS-CoM interaction is dependent on the presence of HS-CoB in MCRred2, we propose that HS-CoB is not only required as a second substrate,16 but also to induce a change forcing the real substrate, CH3-S-CoM, and Ni(I) of the prosthetic group to interact in the active enzyme MCRred1. Acknowledgment. This work was supported by the Swiss National Science Foundation, by the Max Planck Society, and by the Fonds der Chemischen Industrie. Supporting Information Available: Analytical data of [2-33S]coenzyme M (4, NH4+-form) and experimental procedures (PDF). This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Thauer, R. K. Microbiology 1998, 144, 2377-2406. (2) Grabarse, W.; Mahlert, F.; Duin, E. C.; Goubeaud, M.; Shima, S.; Thauer, R. K.; Lamzin, V.; Ermler, U. J. Mol. Biol. 2001, 309, 315-330. (3) Goubeaud, M.; Schreiner, G.; Thauer, R. K. Eur. J. Biochem. 1997, 243, 110-114. (4) Telser, J.; Davydov, R.; Horng, Y. C.; Ragsdale, S. W.; Hoffman, B. M. J. Am. Chem. Soc. 2001, 123, 5853-5860. (5) Tang, Q.; Carrington, P. E.; Horng, Y. C.; Maroney, M. J.; Ragsdale, S. W.; Bocian, D. F. J. Am. Chem. Soc. 2002, 124, 13242-13256. (6) Mahlert, F.; Bauer, C.; Jaun, B.; Thauer, R. K.; Duin, E. C. J. Biol. Inorg. Chem. 2002, 7, 500-513. (7) Mahlert, F.; Grabarse, W.; Kahnt, J.; Thauer, R. K.; Duin, E. C. J. Biol. Inorg. Chem. 2002, 7, 101-112. Erratum: J. Biol. Inorg. Chem. 2002, 7, 351. (8) Duin, E. C.; Cosper, N. J.; Mahlert, F.; Thauer, R. K.; Scott, R. A. J. Biol. Inorg. Chem. 2002, 8, 141-148. (9) Finazzo, C.; Harmer, J.; Jaun, B.; Duin, E. C.; Mahlert, F.; Thauer, R. K.; Van Doorslaer, S.; Schweiger, A. J. Biol. Inorg. Chem., in press. (10) [33S8]-S8 was purchased from Campro Scientific, Berlin, Germany. The analysis of (isotopic) purity of [33S8]-S8 was established by the Kurchatov Institute, Moscow, Russia (isotopic purity: 99.79%; purity: > 99.95%). (11) Castiglioni, A. Gazz. Chim. Ital. 1933, 63, 171. (12) During development of the procedure with natural abundance sulfur, each intermediate was isolated and characterized. The product was converted to the ammonium salt by ion exchange with Amberlite, and inorganic salts (NH4Br) were removed by precipitation of 4 with diethyl ether from a concentrated solution in methanol. The sample used for the experiments with the enzyme was recrystallized from methanol/2-propanol to remove a small inorganic impurity that quenched the MCRred2 signal. The details of the development of the synthesis and the analytical data for all intermediates will be published elsewhere. (13) Gromov, I.; Shane, J.; Forrer, J.; Rakhmatoullin, R.; Rozentzwaig, Yu.; Schweiger, A. J. Magn. Reson. 2001, 149, 196-203. (14) Schweiger, A.; Jeschke, G. Principles of Pulse Electron Paramagnetic Resonance; Oxford University Press: New York, 2001. (15) Morton, J. R.; Preston, K. F. J. Magn. Reson. 1978, 30, 577-582. (16) Horng, Y. C.; Becker, D. F.; Ragsdale, S. W. Biochemistry 2001, 40, 12875-12885.

JA0344314

J. AM. CHEM. SOC.

9

VOL. 125, NO. 17, 2003 4989